The Relevance of Upstream and Downstream Regions of mRNA in the Prediction of Translation Initiation Site of the Protein

Abstract

The correct prediction of Protein Translation Initiation Site is a problem of considerable interest in the molecular biology. In the literature, some papers investigate conservative positions of the mRNA that favor an ideal context for begin of translation, as it is the case of the work of Kozak. The Kozak Consensus is known to reveal some important mRNA \textit{upstream} region positions to recognize an early translation. This work investigated the real contribution of these conservative positions, besides analyzing sequences in the \textit{downstream} region to be considered as non-initiators of translation. In this paper, we use the Support Vector Machine classifier and three different scenarios. The results indicate that the proposed methodology was efficient, guaranteeing a gain in the performance of the larger organisms. For the smaller organisms, the three scenarios investigated maintain the same behavior. However, due to the lack of consideration of the \textit{upstream} region of the mRNA, the proposed methodology allows evaluating organisms not considered in other works, as is the case of the \textit{Caenorhabditis elegans} organism.